git clone https://caio.co/de/chuva

Activity

Cleanup old data with a pattern based on latest file 💬 by Caio 2 months ago

Ensemble dataset has (obviously) a different name. Filled up the disk,
couldn't download newer data.

First outage 🎉🎉

info: expose current head as version by Caio 2 months ago

Remove leftover file 💬 by Caio 2 months ago

Got copy-renamed to moros.rs and I forgot to git-rm

Record lastTick after content swap 💬 by Caio 2 months ago

Just to that I updates correctly when I switch tabs like a madman

No need to `use ndarray::` here by Caio 2 months ago

Add chuva/bin/check to verify loading 💬 by Caio 2 months ago

Load a given dataset using code with manual offsets and compare
vs using ndarray.

Fix ensemble dataset loading 💬 by Caio 2 months ago

It's row-major, not column-major (i.e. the LAST dimension is contiguous,
not the first). RIP muh cpu cache

Add a little helper to diff datasets by Caio 2 months ago

Make download-dataset a bit smarter 💬 by Caio 2 months ago

Take dataset from args. Figures out which PID to terminate through
systemd so I can make multiple instances of the same binary without
worry

Now https://beta.chuva.caio.co is running with the ensemble model

No need for load feature anymore 💬 by Caio 2 months ago
```
Fewer moving pieces. Finally.
```

Chuva

Rain prediction for The Netherlands

Live: https://chuva.caio.co
Demo: https://chuva.caio.co/demo
Git Repository: https://caio.co/de/chuva/

History

Earlier this year (2025) I got fed up with how slow and full of ads the service I used before got, then I found out that KNMI publishes precipitation predictions so I quickly hacked something that dumped what wanted to the terminal

For months, this is all that chuva was: a script downloading the most recent dataset and me ssh'ing into the server to call the program. Then came October with a lot of random rain that barely gets you wet which prompted me to make it easier to access it.

I kept the support for plain-text predictions since I like to avoid going to the browser to prevent distractions. You can use it too if you like, either via the accept header or a txt=1 query string:

$ curl -H accept:text/plain https://chuva.caio.co/demo
It's 10:48

▃▄ ▄▆▆▅▁          ▁▄▅▄▂
^
- Rain ends in 36 minutes
- Rain from 12:15 until 12:40

Project Layout

In alphabetical order:

caveman: A hyper-based, serde-less http1 thing that can answer traditional web requests and knows to shutdown gracefully
chuva: An API for interacting with the KNMI dataset. It loads the dataset to RAM in a format that makes it super easy to answer the "what's the forecast for a given coordinate" question
download-dataset: The "cron job" that retrieves the most recent dataset
etc/systemd: A skeleton of the systemd unit files that drive the service
moros: The web server, front-end for chuva.caio.co. Designed to be killed/restarted periodically to pick up the most recent data

License

This software is licensed under the European Union Public License (EUPL) v. 1.2 only